Lifetime Enhancement of Visible Light Induced Photocharges in

28 mins ago - To evaluate these added benefits, the TiO2:WN thin films were characterized by different techniques. X-ray diffraction patterns and X-ra...
0 downloads 9 Views 5MB Size
Article Cite This: J. Phys. Chem. C XXXX, XXX, XXX−XXX

pubs.acs.org/JPCC

Lifetime Enhancement of Visible Light Induced Photocharges in Tungsten and Nitrogen in situ Codoped TiO2:WN Thin Films N. Delegan,† R. Pandiyan,† S. Johnston,‡ A. Dirany,§ S. Komtchou,§ P. Drogui,§ and M. A. El Khakani*,† Centre-Énergie, Matériaux et Télécommunications, Institut National de la Recherche Scientifique, 1650 Blvd. Lionel-Boulet, Varennes, QC J3X-1S2, Canada ‡ National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, Colorado 80401-3393, United States § Centre-Eau, Terre et Environnement, Institut National de la Recherche Scientifique, 490 Rue de la Couronne, Québec City, QC G1K-9A9, Canada †

ABSTRACT: We report on one-step in situ codoped TiO2 thin films synthesized by cosputtering. The purpose of this acceptor− donor passivated codoping approach is to overcome the optoelectronic limitations that arise for monodoped TiO2 in photocatalytic applications. To evaluate these added benefits, the TiO2:WN thin films were characterized by different techniques. X-ray diffraction patterns and X-ray photoelectron spectral analysis revealed that both N and W dopants are mostly present in the desired substitutional locations. Additionally, the codoping approach was found to reduce the internal strain and defect density of the TiO2:WN films as compared to their monodoped TiO2:N counterparts. This defect reduction is confirmed via photocharge lifetime variation obtained using visible light flash photolysis time-resolved microwave conductivity measurements (FP-TRMC). Photocharge lifetime analysis indicated the presence of three distinct decay processes: charge trapping, recombination, and surface reactions. These characteristic lifetimes of the codoped TiO2:WN films (i.e., 0.08, 0.75, and 11.5 μs, respectively) were found to be about double those of their nitrogen monodoped TiO2:N counterparts (i.e., 0.03, 0.35, and 6.8 μs), quantitatively confirming the effective passivating outcome of the tungsten−nitrogen codoping approach developed here. The practicality of this method was confirmed by integrating the TiO2:WN films as photoanodes for the electro-photocatalytic, solar light driven degradation of a real pollutant (i.e., atrazine). A significant increase in the degradation kinetics, leading to a 4-fold increase in the pseudo-first-order degradation constant for the optimally doped TiO2:WN photoanodes (0.106 min−1) from the undoped TiO2−x ones (0.026 min−1), is a direct consequence of the increased photocharge lifetimes in tandem with visible light photosensitivity.



visible spectrum (∼2.3 eV).3,8,11 However, this narrowing induces defects states that tend to decrease the per-photon photocatalytic efficiency of the material by lowering the photocharge mean free path and increasing their recombination rates.10,12−18 Many studies have theoretically investigated possible pathways to circumvent these undesirable effects induced by Ndoping of TiO2 films. The most promising approach is based on acceptor−donor codoping that would suppress the charge disparities via local electronic passivation, while maintaining visible light photosensitization.19−23 Dopant pairs such as (C +W), (V+N), (Nb+N), and (W+N) have all been proposed as possible candidates that would maintain proper band alignment required for active radical production in aqueous environment under solar illumination.20,21,24,25 Among these pairs, tungsten

INTRODUCTION Titanium dioxide (TiO2) based photocatalysts continue to be extensively studied due to their high photoreactivity in water splitting reactions and in decomposing environmental pollutants. This interest is compounded by the material’s availability, stability, and energy band positioning.1−5 Fundamentally, TiO2 is an n-type semiconductor with a wide intrinsic bandgap (Eg). The exact Eg value depends on the polymorph, with rutile (R) and anatase (A) having a 3.0 and 3.2 eV band gap, respectively. Its large Eg limits the use of pure TiO2 driven photocatalytic applications to the UV portion of the solar spectrum, which represent only ∼4% of the total energy flux at AM1.5G.1−5 This limitation has led to substantial research efforts aimed at narrowing the Eg of the material to photosensitize it in the visible. To this end, various methods (such as self-doping, dye sensitization, cationic and/or anionic doping, etc.2−4,6−10) have been attempted. Substitutional nitrogen (N) doping has shown promising results as the formation for discrete N 2p acceptor states within the TiO2 Eg has been reported to photosensitize the material well into the © XXXX American Chemical Society

Received: November 14, 2017 Revised: January 24, 2018

A

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

of 8.8 W/cm2 for the TiO2 target and a varying RF power density for the W target. Prior to deposition, the chamber was cryo-pumped to a base pressure of 2 × 10−8 Torr. High-purity Ar (99.999%) and N2 (99.995%) gases were then introduced into the chamber. The gas flow rates were monitored to maintain a constant pressure of 1.45 mTorr in the chamber during the sputter-deposition process. To control the N incorporation in the TiO2:WN films, the relative nitrogen mass flow rate ratio RN2 (i.e., [N2]/([N2] + [Ar])) was varied between 0 and 15%. The W incorporation was controlled by varying the W target power density (Wpower) from 0.06 to 0.25 W/cm2. The operating parameters to obtain desired N and W doping concentrations were set on the basis of the optimization work done previously on TiO2:N and TiO2:WN films.7,8 The TiO2:WN films were concomitantly deposited onto silicon, quartz, and Ti substrates. These were mounted on a holder located off-axis at a distance of 20 cm from the sputtering targets and heated during deposition by a quartz lamp heater to an on-substrate temperature of ∼470 °C. Prior to film deposition, the target was systematically sputter-cleaned with Ar ions for ∼15 min with the shutters closed. The thickness of the TiO2:WN films was in situ monitored by means of a calibrated quartz-crystal microbalance and ex situ measured through cross-sectional scanning electron microscopy (SEM) by means of a Jeol JSM-6300F microscope. The thicknesses of all films was of about 300 nm. No post acceleration voltage was intentionally applied to the substrates during the sputterdeposition process (they were nonetheless subjected to a builtin plasma sheath bias of ∼−14 V during their growth). The sputtering based doping process is highly reproducible (as this is one of the inherent advantages of such physical deposition technique); it allows a very tight control of the doping level through the fine adjustment of the various deposition parameters (power, pressure, bias, substrate temperature, gas flow ratio, etc.). Indeed, many samples for each doping conditions were deposited (in different deposition batches), and the initially targeted doping concentrations were always achieved almost perfectly within the precision of the XPS measurements (which is estimated to 0.1 at. %). The crystalline structure of the films on quartz substrates was characterized by means of a PANalytical X-Pert Pro X-ray diffractometer (XRD) system using Cu Kα radiation of 1.5418 Å. The atomic composition and chemical bonding states of the films were systematically investigated by means of X-ray photoelectron spectroscopy (XPS). High-resolution XPS core level spectra of all the samples were collected by using the ESCALAB 220i-XL spectrophotometer (Thermo VG Scientific Ltd.), equipped with a monochromatized Al Kα (1486.6 eV) twin-anode source after a systematic in situ surface cleaning by means of 5 keV Ar+ ion sputtering gun with an average ∼0.15 mA current. The obtained spectra were fitted using the CasaXPS ver. 2.3.15 software with Shirley background approximation. Surface charge effects were controlled for by verifying the location of the presputter cleaned C 1s peak (C− C bond). All atomic species were fitted by using fixed values for intrinsic properties such as peak shape (i.e., GL(30)), and the doublet area/separation relationships taken from the NIST Standard Reference Database 20, Version 4.1 (https://srdata. nist.gov/xps/). The atomic concentrations were evaluated by using their relative sensitivity factors (RSF) defined in the NIST databases. Given the overlapped nature of the tungsten spectra, both the W 4f and W 4d (not shown) binding energy windows were used to ensure appropriate quantification. FP-

(W) and nitrogen (N) codoping has been outlined as a candidate of choice.19−21,23 Some experimental evidence of the WN-codoping approach has been recently put forth. On one hand, Cui et al.26 and Thind et al.27,28 independently showed that codoping has led to a favorable Eg reduction, with increased degradation kinetics as compared to the case for TiO2:N samples. On the other hand, Kubacka et al. demonstrated similar results where they have clearly observed an improvement of optoelectronic properties of TiO2:WN and an easier formation of −OH radicals at the photoanode surface.29 However, in all cases, the exact optoelectronic properties and mechanisms have yet to be elucidated. Folli et al.30 and Bloh et al.31 reported that W and N incorporation (via wet chemistry synthesis routes) altered the band gap structure of the TiO2:WN films. Unfortunately, most of their dopants were in interstitial locations, resulting in the formation of undesired deep gap states. In other words, practical demonstration and understanding of the beneficial substitutional type acceptor−donor passivated codoping of TiO2:WN films is still lacking. Therefore, the main purpose and motivation of this study is to practically investigate and optimize codoping strategies to overcome the limitations associated with monodoping. In particular, insights into the per-photon efficiency of TiO2 based photoanodes through the determination of their photocharge lifetimes and associated mean-free paths is of paramount interest. In this context, this study reports on a systematic investigation of photocharge lifetimes and their variation as a function of various doping schemes of sputtered TiO2 films. We used visible light flash photolysis time-resolved microwave conductivity measurements (FP-TRMC) to study the photocharge decay upon illumination of different TiO2 photoanodes, namely undoped (TiO 2−x ), monodoped (TiO2:N and TiO2:W), and codoped (TiO2:WN) films as a function of dopant incorporation and varying doping levels. The photoanodes were prepared using a one-step reactive magnetron sputtering process. These were subjected to different characterization techniques including scanning electron microscopy (SEM), X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS) to qualify their structure, morphology, and chemical composition as a function of their doping states. In parallel, FP-TRMC was used to ascertain the effects of various doping configurations of the sputtered TiO2 films on their optoelectronic behavior, namely, their characteristic photocharge lifetimes. We were thus able to demonstrate the much longer lifetimes exhibited by the optimally codoped TiO2:WN, confirming thereby the occurrence of the acceptor− donor passivation. Finally, the application of these differently doped photoanodes to the electrophotocatalytic (EPC) degradation of a water pollutant (i.e., atrazine) confirmed that the optimally codoped TiO2:WN photoanodes are the most effective as compared to other doped TiO2 films. In fact, we were able to establish a direct correlation of the fundamental optoelectronic properties observed via TRMC with results obtained while using the TiO2:WN films as photoanodes in the electro-photocatalytic degradation of a pesticide, atrazine under solar illumination.



EXPERIMENTAL SECTION The TiO2:WN films were deposited from the cosputtering of both a 3 in. diameter TiO2 target (99.995% purity) and a 3 in. diameter W target (99.95% purity) using RF (13.56 MHz) magnetron sputter guns operating at a constant power density B

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C TRMC measurements were performed using a microwave generator (μwave) operating at either 7 or 20 GHz as a probe for the photoconductive decay in the sample cavity. The pulsed light excitation was produced using an yttrium aluminum garnet (YAG) optical parametric oscillator to provide light pulses at 10 Hz with a wavelength of 420 nm and an intensity of ∼1 mW/ cm2. The pulses were 3−5 ns wide with a ∼5 ns detector response time. The photogenerated carriers modulate the μwave transmission coefficient ,which, in turn, is transformed into a change in voltage using a PIN diode.



RESULTS AND DISCUSSION All of the sputter-deposited TiO2 films were systematically examined via SEM. These were all found to exhibit a low surface roughness with no apparent porosity. Figure 1 shows a

Figure 1. Typical cross-sectional SEM micrograph of the sputterdeposited TiO2:WN films. The image shown here is of the TiO2:WN #2 sample, 250 nm thick with the front facing face exposed by cleaving of the Si substrate. Figure 2. Typical XRD spectra of the sputter-deposited TiO2, TiO2:N, TiO2:W, and TiO2:WN thin films. The anatase (A) and rutile (R) phases are identified.

typical cross-sectional SEM micrograph of a ∼250 nm thick TiO2:WN thin film, where a dense morphology along with a smooth surface can be clearly seen. No differences in morphology were observed between the TiO2 samples of various doping schemes (i.e., TiO2−x, TiO2:W, TiO2:N, and TiO2:WN). From cross-sectional SEM images, we were able to determine the actual film thickness for the various films (all films are in the ∼200−400 nm range). The crystallinity of our codoped TiO2:WN thin films was investigated through systematic XRD measurements as a function of the doping scheme. Figure 2 presents elected XRD spectra to illustrate the main differences between the undoped, N-doped, W-doped, and WN-codoped TiO2 films. Typically, at deposition temperatures of ∼475 °C, the anatase phase is formed.7 However, in the case of TiO2−x films, we know that they crystallize in the rutile (R) polymorph with its characteristic (110), (101), (200), (111), (210), (211), and (220) diffraction peaks present (identified from left to right; JCPDS card #88−1175). In fact, this is an expected result of sputter depositing TiO2 near the anatase−rutile transition point (∼500−600 °C) if no oxygen is introduced during the sputtering (preferential sputtering of O atoms causes oxygen vacancy (VO) formation7,32). The presence of oxygen vacancies enhances light absorption of TiO2−x at ∼420 nm due to VO defect bands just below the CBM, a requirement for visible light excitation for the TRMC experiments. In the case of

TiO2:N, we note the sole presence of the anatase polymorph with its characteristic (101), (004), and (200) peaks (identified from left to right; JCPDS card #21−1272). The N atoms are expected to fill out a sufficient amount of VO to favor the formation of the anatase phase. Interestingly, the coexistence of both A and R phases is observed upon W doping. W atoms can only substitute for Ti ones (due to size constraints)33 and therefore cannot fill oxygen vacancies. Tungsten is, however, expected to increase the VO formation energy,22,23 resulting in both phases being formed. Finally, when W and N dopants are introduced in the TiO2 structure, a well-defined anatase phase is formed with an estimated crystallite size of about ∼25 nm (determined using Williamson-Hall analysis and Scherrer formula approximation). This indicates that both dopants work synergistically at the lattice level not only to compensate for oxygen vacancies but also to reduce the strain, most likely because of local charge compensation. To characterize the chemical bonding states of the sputtered films, XPS analyses of the N 1s, Ti 2p, O 1s, W 4f (this is also the Ti 3p window), and W 4p core level spectra were performed. After appropriate deconvolution of the XPS spectra for all studied samples (Table 1), the various doping-induced changes in the chemical states present in a typical TiO2:WN C

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Table 1. X-ray Photoelectron Spectroscopy Derived Dopant Atomic Concentrations (at. %), X-ray Diffraction Determined Dominant Crystalline Structures, and a First-Order Approximation Factor ([Ti3+] + [Ti2+])/[Titotal] for the VO Densitya name TiO2−x TiO2:N #1 TiO2:W #1 TiO2:W #2 TiO2:W #3 TiO2:WN #1 TiO2:WN #2 TiO2:WN #3 TiO2:WN #4 a

nitrogen (at. %) 0 5.8 0 0 0 6.4 7.7 9.1 7.7

± 0.1

± ± ± ±

0.1 0.1 0.1 0.1

tungsten (at. %) 0 0 0.7 2.2 3.1 0.8 1.6 2.5 2.6

± ± ± ± ± ± ±

0.1 0.1 0.1 0.1 0.1 0.1 0.1

crystallinity

([Ti3+] + [Ti2+])/[Titotal]

rutile anatase anatase & rutile anatase & rutile amorphous amorphous amorphous anatase anatase

0.30 0.48 0.21 0.39 0.35 0.45 0.40 0.28 0.39

The error on the XPS contents represents the smallest detectable atomic concentration under the experimental conditions used.

Figure 3. High-resolution XPS spectra of the O 1s, Ti 2p, N 1s, and W 4f core levels (from left-to-right) of the TiO2:WN #2 thin film. The Ti 3p deconvoluted states are dashed to prioritize visibility of the W 4f states. For display purposes, normalized counts are used for the vertical axis.

codoped film are shown in Figure 3 (TiO2:WN #2 sample used for display purposes). For visual simplicity, all of the doping schemes can be seen in the TiO2:WN codoped sample with no hampering overlap. Additionally, the Ti 3p core-level energy signatures in the W 4f and Ti 3p shared window are dashed to visually lighten the figure. For the TiO2−x films, the Ti 2p3/2 peak can be decomposed into two main components corresponding to two oxidation states of Ti, namely, Ti4+ and Ti3+ at 458.6 and 456.9 eV, respectively. Their corresponding Ti 2p1/2 energy levels can be found at 5.7 and 5.5 eV higher binding energy (BE) with a 2p1/2:2p3/2 branching ratio of 1:2. These can also be observed in the Ti 3p spectra with the main Ti4+ 3p peak at 37.2 eV and the T3+ and Ti2+ at 1.30 and 2.27 eV lower BE, respectively.7,33−36 The presence of Ti3+ is due to the formation of the 2Ti3+−VO″ defect association pair, as a consequence of preferential sputtering of the TiO2 target.7,32 These states can be mirror derived from the O 1s spectra with the main OA peak at 529.9 eV assigned to lattice oxygen in the TiO2 matrix and the OB|OC peaks assigned to substoichiometric lattice oxygen species (in Ti3+ and Ti2+ lattice environments, respectively) each at subsequently ∼0.5 eV higher binding energy.34,35 Additionally, we note the presence of a fourth component, denoted as Odef at 530.7 eV, which is typically associated with defective surface oxygen and surface hydroxide species.34,35,37 Upon nitrogen incorporation in the films, the N 1s peak is seen to exist in two characteristic components: Nsub and Nint at 396.2 and 397.2 eV, respectively. The Nsub peak is known to arise from substitutional −Ti−N− type bonds, and the Nint peak is most likely interstitial −O−N type bonds within the titania crystal structure.3,6,7,10,13,14 The introduction of N into the titania is accompanied by the appearance of Ti2+

oxidation states in the Ti 2p spectra (Ti2+ 2p3/2 at 455.3 eV and the 2p1/2 component at 460.9 eV BE).3,38 These reduced valence states are typically associated with electronic defects originating from heavily substoichiometric lattice structures.3,4,7,15,39 Concomitant sputtering of a W target in tandem with the TiO2 target was used for W doping. This resulted in the presence of two main features in the W 4f BE window. These are associated with the W6+ and W4+ states (in the oxide form) with their main 4f7/2 components at 34.5 and 32.3 eV, respectively. Their corresponding 4f5/2 doublets are found at 2.2 eV higher BE with a 3:4 branching ratio.33,40−43 The TiO2:WN films exhibit W and N binding energy signatures that combine the XPS features of each monodoped sample accordingly. Both the W and N dopants can be associated with mainly substitutional type incorporation replacing Ti and O atoms, respectively. Finally, using the ([Ti3+] + [Ti2+])/ [Titotal] ratio as an indicator of the VO density, the XPS analysis reinforces the XRD observations, indicating that codoped films have lower defect densities as compared to their N-doped counterparts. A summary of the XPS quantitative dopant content along with the corresponding TiO2 polymorph determined by XRD for all the TRMC tested films is presented in Table 1. This is accompanied by the relative area ratio ([Ti3+] + [Ti2+])/[Titotal] for the studied samples. We note that the sample with 9.1 at. % nitrogen and 2.5 at. % tungsten shows the lowest approximated VO density. To better understand the effect of WN-codoping on the photoactivity of TiO2 thin films, a detailed investigation of the charge carrier dynamics was necessary. To this end, FP-TRMC measurements were carried out to probe the time-resolved charge in conductivity σ(t) induced by pulsed laser excitation of D

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

three characteristic time constants, namely, τtrap, τrec, and τsurf (with their respective time scales on the order of 0.05, 1, and 10 μs).4 Thus, τtrap, τrec, and τsurf characterize charge trapping (e−| h+ → e−tr|h+tr), charge recombination (e− + h+|e−tr + h+|e− + h+tr → hυ), and surface reactions (e− + O2 → O2−), respectively.4 Figure 5 presents the extracted characteristic

undoped, monodoped, and codoped TiO2 films. During laser excitation, the change in microwave power transmission (ΔP(t)) from the sample cavity as a function of time is recorded. It is proportional to the change in photoconductance (Δσ(t)) of the sample by a sensitivity factor A, which is determined by the known dimensions and characteristics of the sample cavity and hardware used, according to the following relation:18,44−47 ΔP(t)/P(t) = A·Δσ(t). Thus, with Δσ(t) ∼ μ·ΔN (where μ is the charge mobility, and ΔN the charge population change), the time evolution of ΔP(t)/P(t) gives information about the recombination or trapping of charge carriers (further theoretical and experimental explanations can be found elsewhere18,44−47). Figure 4 shows typical TRMC

Figure 5. Photocharge lifetimes for all three characteristic decay lifetimes (τtrap, τrec, and τsurf) as a function of doping scheme.

time constants of the different TiO2, TiO2:N, TiO2:W, and TiO2:WN photoanodes. We note that there is a clear dependence of the photocharge lifetimes with the doping scheme, with the W-doped samples showing the highest overall lifetimes (this will be discussed further). Interestingly, in the case of WN-codoped thin films, we note an almost order of magnitude variation among samples with various dopant loadings. These characteristic lifetimes are presented in Table 2.We can note that total dopant incorporation efficiencies are insufficient in highlighting a relationship between codoping and photocharge lifetimes.

Figure 4. Representative FP-TRMC signals of the TiO2−x, TiO2:N, TiO2:W, and TiO2:WN thin films (doping concentrations of the samples presented can be found in the figure legend). These were excited with a 5 ns wide, 420 nm laser pulse. The prepulse baseline was ∼10−2 mV.

decay signals for the TiO2, TiO2:N, TiO2:W, and TiO2:WN films following their illumination by a 5 ns pulse duration of the laser emitting at 420 nm. One can note a clear difference in the signal amplitude for the different doping schemes. It is known that the ΔP(t)/P(t) of charge carriers in undoped and doped TiO2 films is not directly proportional to the excitation light intensity (Iex).18,48 This change in intensity ΔP(t)/(Iex·P(t)) is typically affected by three main factors: (i) trap filling effects at low excitation energy, (ii) recombination effects at high excitation energy, and (iii) dopant induced donor and/or acceptor populations,18,44−48 all of which can be at work here. Here, focus is put on the extraction of lifetime data from the decay curves of Figure 4, as they depend purely on the time variation of the signal and not the overall signal intensity. To derive the photocharge lifetimes from the TRMC data, the A·Δσ(t) traces were fitted using a triple exponential function, representative of the three decay processes that affect the charge carriers in the material. Thus, A·Δσ(t) can be written as A·Δσ(t) = ∑An·exp(−t/τn), where An and τn are the respective proportionality and the time constants of each of the three decay processes. In other words, all the lifetime curves can be separated into three decay regimes that were associated with

Table 2. Extracted Characteristic Photocharge Lifetimes (from FP-TRMC Fittings) for the WN Codoped Samples τtrap (μs)

name TiO2:WN TiO2:WN TiO2:WN TiO2:WN

#1 #2 #3 #4

0.058 0.037 0.077 0.092

± ± ± ±

0.021 0.016 0.015 0.007

τrec (μs) 0.59 0.26 0.75 0.59

± ± ± ±

0.13 0.07 0.13 0.40

τsurf (μs) 9.9 3.5 11.5 7.4

± ± ± ±

1.9 0.3 1.2 0.4

In an effort to correlate the obtained time constant values with the film composition and bonding states of the dopants, we have systematically analyzed the XPS spectra for the different TiO2 doping schemes. Table 3 summarizes the XPS derived concentration of each dopant along with its incorporation state (i.e., substitutional/interstitial for nitrogen, and W6+/W4+ for tungsten). Passivated codoping models mainly focus on W6+ and Nsub species.19−21,23 W4+ is assumed to have little to no local electronic effect upon substituting for a E

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

TiO2:N, we see that the lifetimes are greatly hindered by the local charge defects brought up by the Nsub charge disparity forcing the formation of VO. In the case of TiO2:WN codoped films, the W6+ species are partially balancing the electronic disparity brought on by Nsub, allowing the lifetimes to converge closer to the undoped TiO2−x. Finally, in the case of TiO2:W samples, the W6+ seem to be mitigating the effects of naturally occurring VO (i.e., we can directly observe the predicted increase in VO formation energy as a consequence of W6+ incorporation22,23). These results clearly indicate that codoping with a suitable acceptor−donor pair, such as W and N, in turn increases the photocharge lifetimes significantly via suppression of the 2Ti3+−VO″ defect association states. This enables the material to maintain a high photosensitivity (matching that of optimally doped TiO2:N with Eg ∼ 2.3 eV), while increasing its associated photocharge lifetimes to values comparable to those of TiO2−x (∼0.27, 1.9, and 12.2 μs for τtrap, τrec, and τsurf, respectively). On the basis of these observations, we would predict that TiO2:WN should perform better than any W or N monodoped titania when used as photoanodes in a visible light electrophotocatalytic processes. To assess the validity of such a prediction, the same TiO2 thin films studied here were deposited onto Ti grids (as large as 6 in. in diameter) and used as photoanodes in a 1 L home built EPC reactor, details of which are presented elsewhere.49 The performance of our photoanodes was directly evaluated toward the degradation of an agricultural pollutant, namely atrazine. Atrazine is a potent endocrine disruptor widely used in North America as a pesticide/herbicide. Critically, studies have reported the persistence of atrazine in surface, ground, and drinking water.50,51 This problem is compounded by the fact that photolysis of atrazine is only possible with high-energy UV radiation (dechlorination mechanism).52 In other words, there is an urgent need for a low-cost, environmentally friendly approach for the degradation of atrazine by using readily available sunlight. Atrazine was selected as a relevant electrophotocatalytic testbed to prove the effectiveness of the codoped TiO2:WN photoanodes under sunlight. Further details on the subject can be found elsewhere.52 For the purpose of the present study, synthetic solutions containing atrazine concentrations of 60 μg/L (60 ppb, as these are typical concentrations of heavily polluted areas) were prepared and treated in our EPC reactor under AM1.5G solar simulator while continuously circulating them with treatment times reaching up to 6 h. The exact EPC degradation mechanisms are elaborated upon elsewhere.52 Figure 7 shows typical time-dependent degradation curves of atrazine by the different photoanodes studied here (i.e., TiO2−x, TiO2:N, TiO2:W, and TiO2:WN). One can note that all the photoanodes successfully degraded atrazine after 3 h of treatment time, allowing the residual pollutant concentrations to reach 1−2 ppb (the environmentally acceptable limit). However, one can note that the kinetics of atrazine degradation heavily depends on the photoanode nature. Indeed, although it takes ∼3 h for the EPC treatment of atrazine to reach the ∼1.5 ppb threshold using a TiO2−x based photoanode, this time is reduced to ∼45 min with a TiO2:WN based photoanode. The pseudo-first-order degradation kinetic constants for each studied photoanode were extracted and are presented in Table 4. It is clearly seen that compared to the case for TiO2−x, TiO2:N doping doubles the rate of atrazine degradation, confirming that the photosensitization (effective reduction of the bandgap size) permits

Table 3. XPS Derived Quantifications for Each Dopant Subspeciesa name TiO2−x TiO2:N #1 TiO2:W #1 TiO2:W #2 TiO2:W #3 TiO2:WN #1 TiO2:WN #2 TiO2:WN #3 TiO2:WN #4

Nsub (at. %) 0 4.2 0 0 0 4.1 4.6 5.6 5.8

± 0.1

± ± ± ±

0.1 0.1 0.1 0.1

Nint (at. %) 0 1.6 0 0 0 2.2 3.1 3.5 1.8

± 0.1

± ± ± ±

0.1 0.1 0.1 0.1

W6+ (at. %) 0 0 0.3 0.7 1.1 0.2 0.4 1.1 1.0

± ± ± ± ± ± ±

0.1 0.1 0.1 0.1 0.1 0.1 0.1

W4+ (at. %) 0 0 0.4 1.6 2.0 0.5 1.2 1.4 1.6

± ± ± ± ± ± ±

0.1 0.1 0.1 0.1 0.1 0.1 0.1

a Nsub and Nint denote the substitutional and interstitial type nitrogen, and W6+ and W4+ denote the oxidative species of the W dopant atoms. The statistical error represents the smallest detectable atomic concentration given the experimental conditions.

Ti4+ atom,33 and Nint is typically associated with the formation of −O−N type ions where nitrogen assumes a positive valence.12−14 From these ideas, we can propose that two defects created by N sub incorporation can be locally compensated for by the introduction of a W6+ atom. In other words, we can define an empirical XPS derived acceptor−donor passivation parameter: 2[W6+] − [Nsub.]. This parameter is expected to predict the reduction in defect states brought on via local passivation of charge disparities.19,21 This, in turn, should manifest itself as an increase of photocharge lifetimes through the reduction of VO defect centers that would otherwise form in an uncompensated environment. Consequently, the TRMC variation of photocharge lifetimes as a function of 2[W6+] − [Nsub.] is presented in Figure 6. We clearly note that the introduction of W6+ states increases the photocharge lifetimes considerably. Interestingly, the passivation effect brought in via introduced W6+ applies to both W monodoped and WN codoped samples. In the case of

Figure 6. Variation of photocharge time constants as a function of approximated W6+ passivation of Nsub species (in the case of TiO2:WN) and VO (for TiO2:W). F

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Figure 7. Electrophotocatalytic performance of the best TiO2−x, TiO2:N, TiO2:W, and TiO2:WN photoanodes under AM1.5G light with a 60 ppb initial concentration of atrazine. The inset shows the extracted degradation constants (pseudo-first-order) as a function of W6+ passivation for photoanodes with Eg ∼ 2.3 eV (i.e., TiO2:N #1 and TiO2:WN #2 & #3).

highlight the added value of the passivation, the degradation constants of photoanodes with identical Eg values (around 2.3 eV) were plotted against the passivation parameter (2[W6+] − [Nsub.]) and reported in the inset of Figure 7. Even if we have only 3 points, we can note that there is a marked increase of the degradation efficiency of atrazine with the increase of the passivation parameter associated with longer photocharge lifetimes. This reinforces that for photoanodes with identical photoharvesting capabilities (i.e., similar bandgaps), we obtain an increase in the EPC performance by directly increasing the photocharge lifetimes via passivated acceptor−donor type doping using W and N. Specifically, through a W dopant induced reduction in the 2Ti3+−VO″ defect pair density.

Table 4. Extracted Pseudo-First-Order Degradation Constants for the Photoanodes Evaluated for the EPC Degradation of Atrazine under AM1.5G Illumination with Their Corresponding Eg Values name TiO2−x TiO2:W #2 TiO2:N #1 TiO2:WN #2 TiO2:WN #3

degradation constants (min−1) 0.026 0.057 0.047 0.097 0.106

± ± ± ± ±

0.0017 0.0053 0.0017 0.0027 0.0090

Eg (eV) 3.2 3.0 2.3 2.3 2.3

± ± ± ± ±

0.1 0.1 0.1 0.1 0.1

the use of a greater portion of the solar spectrum for the EPC process. The exact quantification of the bandgap associated with each doping scheme was done and previously reported elsewhere.8 The Eg values for TiO2−x, TiO2:N, TiO2:W, and TiO2:WN films are summarized in Table 4. Interestingly, even if the Eg of TiO2:W films is larger than that of TiO2:N (3.0 eV for the 2.8 at. % W-doped photoanode vs 2.3 eV for the 5.8 at. % N-doped photoanode), their EPC degradation performance was found to be very similar to that of TiO2:N (Figure 7). This is believed to be a direct consequence of the rather longer lifetimes of photocharges in the TiO2:W films, which compensate for their lack of visible light photon sensitivity. In other words, whereas TiO2:N films profit from their increased photosensitivity to the visible spectrum, the TiO2:W samples offer longer lifetimes and therefore increased reaction probability of produced photocharges. When the two phenomena are combined in the codoping scheme, we can observe that these two contributions act synergistically, allowing reaction constants to increase of about 1 order of magnitude, going from 0.026 min−1 for the TiO2−x to 0.106 min−1 for the optimally codoped TiO2:WN photoanodes. To



CONCLUSION In summary, we have demonstrated the effectiveness of using a reliable RF magnetron sputtering process for the in situ controlled doping and codoping of TiO2 thin films, allowing the synthesis of TiO2−x, TiO2:N, TiO2:W, and TiO2:WN films. A partial indicator of acceptor−donor passivation is presented in the form of a relaxation of the crystalline strain when both W and N are incorporated into the structure, allowing the titania to recover a well-defined anatase phase. These results are corroborated by XPS analyses, which revealed the presence of both W and N in mostly substitutional states for the mono- and codoped TiO2 films. However, detailed analysis of the visible light FP-TRMC signals revealed that three regimes are responsible for the photocharge decay: charge trapping (e−|h+ → e−tr|h+tr), charge recombination (e− + h+|e−tr + h+|e− + h+tr → hυ), and surface reactions (e− + O2 → O2−). Typical lifetimes for TiO2−x films for these three regimes were determined to be around 0.05, 1, and 10 μs, respectively. G

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Intended for Electro-Photocatalytic Applications. J. Phys. Chem. C 2016, 120 (1), 631−638. (9) Scanlon, D. O.; Dunnill, C. W.; Buckeridge, J.; Shevlin, S. a; Logsdail, A. J.; Woodley, S. M.; Catlow, C. R. a; Powell, M. J.; Palgrave, R. G.; Parkin, I. P.; et al. Band Alignment of Rutile and Anatase TiO2. Nat. Mater. 2013, 12 (9), 798−801. (10) Di Valentin, C.; Pacchioni, G.; Selloni, A. Origin of the Different Photoactivity of N-Doped Anatase and Rutile TiO2. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 70 (8), 85116. (11) Daghrir, R.; Drogui, P.; Delegan, N.; El Khakani, M. A. Electrochemical Degradation of Chlortetracycline Using N-Doped Ti/ TiO2 Photoanode under Sunlight Irradiations. Water Res. 2013, 47 (17), 6801−6810. (12) Di Valentin, C.; Pacchioni, G.; Selloni, A.; Livraghi, S.; Giamello, E. Characterization of Paramagnetic Species in N-Doped TiO 2 Powders by EPR Spectroscopy and DFT Calculations. J. Phys. Chem. B 2005, 109 (23), 11414−11419. (13) Lynch, J.; Giannini, C.; Cooper, J. K.; Loiudice, A.; Sharp, I. D.; Buonsanti, R. Substitutional or Interstitial Site-Selective Nitrogen Doping in TiO 2 Nanostructures. J. Phys. Chem. C 2015, 119 (13), 7443−7452. (14) Peng, F.; Cai, L.; Yu, H.; Wang, H.; Yang, J. Synthesis and Characterization of Substitutional and Interstitial Nitrogen-Doped Titanium Dioxides with Visible Light Photocatalytic Activity. J. Solid State Chem. 2008, 181 (1), 130−136. (15) Torres, G. R.; Lindgren, T.; Lu, J.; Granqvist, C.-G.; Lindquist, S.-E. Photoelectrochemical Study of Nitrogen-Doped Titanium Dioxide for Water Oxidation. J. Phys. Chem. B 2004, 108 (19), 5995−6003. (16) D’Arienzo, M.; Siedl, N.; Sternig, A.; Scotti, R.; Morazzoni, F.; Bernardi, J.; Diwald, O. Solar Light and Dopant-Induced Recombination Effects: Photoactive Nitrogen in TiO 2 as a Case Study. J. Phys. Chem. C 2010, 114 (42), 18067−18072. (17) Liu, B.; Wen, L.; Zhao, X. The Photoluminescence Spectroscopic Study of Anatase TiO 2 Prepared by Magnetron Sputtering. Mater. Chem. Phys. 2007, 106 (2−3), 350−353. (18) Katoh, R.; Furube, A.; Yamanaka, K.; Morikawa, T. Charge Separation and Trapping in N-Doped TiO 2 Photocatalysts: A TimeResolved Microwave Conductivity Study. J. Phys. Chem. Lett. 2010, 1 (22), 3261−3265. (19) Yin, W.-J.; Tang, H.; Wei, S.-H.; Al-Jassim, M. M.; Turner, J.; Yan, Y. Band Structure Engineering of Semiconductors for Enhanced Photoelectrochemical Water Splitting: The Case of TiO2. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 82 (4), 45106. (20) Ç elik, V.; Mete, E. Range-Separated Hybrid ExchangeCorrelation Functional Analyses of Anatase TiO2 Doped with W, N, S, W/N, or W/S. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 86 (20), 205112. (21) Li, M.; Zhang, J.; Zhang, Y. First-Principles Calculation of Compensated (2N, W) Codoping Impacts on Band Gap Engineering in Anatase TiO2. Chem. Phys. Lett. 2012, 527, 63−66. (22) Long, R.; English, N. J. Synergistic Effects on Band GapNarrowing in Titania by Codoping from First-Principles Calculations. Chem. Mater. 2010, 22 (5), 1616−1623. (23) Long, R.; English, N. J. First-Principles Calculation of NitrogenTungsten Codoping Effects on the Band Structure of Anatase-Titania. Appl. Phys. Lett. 2009, 94 (13), 132102. (24) Wang, D.; Zou, Y.; Wen, S.; Fan, D. A Passivated Codoping Approach to Tailor the Band Edges of TiO2 for Efficient Photocatalytic Degradation of Organic Pollutants. Appl. Phys. Lett. 2009, 95 (1), 012106. (25) Gai, Y.; Li, J.; Li, S.-S.; Xia, J.-B.; Wei, S.-H. Design of NarrowGap TiO2: A Passivated Codoping Approach for Enhanced Photoelectrochemical Activity. Phys. Rev. Lett. 2009, 102 (3), 36402. (26) Cui, X.; Rong, S.; Cao, Y.; Yin, Y.; Li, S.; Li, M. One-Step Hydrothermal Synthesis of Nitrogen and Tungsten Codoped TiO2 Nanorods with High Visible Light Photocatalytic Activity. Appl. Phys. A: Mater. Sci. Process. 2013, 113 (1), 47−51.

Doping with W unilaterally increases the photocharge lifetimes, whereas N doping showed the lowest lifetimes. Interestingly, the synergetic codoping with both W and N was shown to lead to an order of magnitude increase in the photocharge lifetimes, bringing them to about 0.077, 0.75, and 11.5 μs (from 0.03, 0.35, and 6.8 μs, respectively, for the monodoped TiO2:N thin films). It is to be noted that W monodoping provided even longer lifetimes (as long as, 0.95, 7.99, and 125.8 μs, respectively). However, the wider bandgap of the TiO2:W films (∼3.0 eV) limits their capacity to harvest visible light and hence limits their overall EPC performance. This suggests that a balance should be met between narrower bandgap and longer photocharge lifetimes to optimize the EPC degradation performance, as is well demonstrated here, for the case of codoped TiO2:WN photoanodes toward atrazine degradation. Finally, by correlating the atrazine degradation efficiency to the approximated passivation parameter, our results highlight that the presence of W6+ states in the films is important to achieve an effective passivation in synergy with substitutional nitrogen. These studies have practically led to the optimization of the EPC degradation of a real pollutant (atrazine).



AUTHOR INFORMATION

Corresponding Author

*M. A. El Khakani. E-mail: [email protected]. ORCID

N. Delegan: 0000-0002-1240-6409 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors acknowledge the financial support from NSERC (the Natural Science and Engineering Research Council of Canada) and the FRQNT (Le Fonds de Recherche du QuébecNature et Technologies) through its strategic Network “Plasma-Québec”.



REFERENCES

(1) O’Regan, B.; Grätzel, M. A Low-Cost, High-Efficiency Solar Cell Based on Dye-Sensitized Colloidal TiO2 Films. Nature 1991, 353 (6346), 737−740. (2) Khan, S. U. M.; Al-Shahry, M.; Ingler, W. B. Efficient Photochemical Water Splitting by a Chemically Modified N-TiO2. Science 2002, 297 (5590), 2243−2245. (3) Emeline, A. V.; Kuznetsov, V. N.; Rybchuk, V. K.; Serpone, N. Visible-Light-Active Titania Photocatalysts: The Case of N-Doped TiO 2 sProperties and Some Fundamental Issues. Int. J. Photoenergy 2008, 2008, 1−19. (4) Fujishima, A.; Zhang, X.; Tryk, D. TiO2 Photocatalysis and Related Surface Phenomena. Surf. Sci. Rep. 2008, 63 (12), 515−582. (5) Hashimoto, K.; Irie, H.; Fujishima, A. TiO 2 Photocatalysis: A Historical Overview and Future Prospects. Jpn. J. Appl. Phys. 2005, 44 (12), 8269−8285. (6) Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. VisibleLight Photocatalysis in Nitrogen-Doped Titanium Oxides. Science 2001, 293 (5528), 269−271. (7) Delegan, N.; Daghrir, R.; Drogui, P.; El Khakani, M. A. Bandgap Tailoring of in-Situ Nitrogen-Doped TiO2 Sputtered Films Intended for Electrophotocatalytic Applications under Solar Light. J. Appl. Phys. 2014, 116 (15), 153510. (8) Pandiyan, R.; Delegan, N.; Dirany, A.; Drogui, P.; El Khakani, M. A. Probing the Electronic Surface Properties and Bandgap Narrowing of in Situ N, W, and (W,N) Doped Magnetron-Sputtered TiO 2 Films H

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C (27) Thind, S. S.; Wu, G.; Tian, M.; Chen, A. Nanomaterials for Promising Environmental Applications. Nanotechnology 2012, 23 (47), 475706. (28) Thind, S. S.; Wu, G.; Chen, A. Synthesis of Mesoporous Nitrogen−tungsten Co-Doped TiO2 Photocatalysts with High Visible Light Activity. Appl. Catal., B 2012, 111−112, 38−45. (29) Kubacka, A.; Bachiller-Baeza, B.; Colón, G.; Fernández-García, M. Doping Level Effect on Sunlight-Driven W,N-Co-Doped TiO2Anatase Photo-Catalysts for Aromatic Hydrocarbon Partial Oxidation. Appl. Catal., B 2010, 93 (3−4), 274−281. (30) Folli, A.; Bloh, J. Z.; Beukes, E.-P.; Howe, R. F.; Macphee, D. E. Photogenerated Charge Carriers and Paramagnetic Species in (W,N)Codoped TiO 2 Photocatalysts under Visible-Light Irradiation: An EPR Study. J. Phys. Chem. C 2013, 117 (42), 22149−22155. (31) Bloh, J. Z.; Folli, A.; Macphee, D. E. Adjusting Nitrogen Doping Level in Titanium Dioxide by Codoping with Tungsten: Properties and Band Structure of the Resulting Materials. J. Phys. Chem. C 2014, 118 (36), 21281−21292. (32) Brassard, D.; El Khakani, M. a.; Ouellet, L. Substrate Biasing Effect on the Electrical Properties of Magnetron-Sputtered High-K Titanium Silicate Thin Films. J. Appl. Phys. 2007, 102 (3), 034106. (33) Sathasivam, S.; Bhachu, D. S.; Lu, Y.; Chadwick, N.; Althabaiti, S. a.; Alyoubi, A. O.; Basahel, S. N.; Carmalt, C. J.; Parkin, I. P. Tungsten Doped TiO2 with Enhanced Photocatalytic and Optoelectrical Properties via Aerosol Assisted Chemical Vapor Deposition. Sci. Rep. 2015, 5, 10952. (34) Biesinger, M. C.; Payne, B. P.; Grosvenor, A. P.; Lau, L. W. M.; Gerson, A. R.; Smart, R. S. C. Resolving Surface Chemical States in XPS Analysis of First Row Transition Metals, Oxides and Hydroxides: Sc, Ti, V, Cu and Zn. Appl. Surf. Sci. 2010, 257 (7), 887−898. (35) Jackman, M. J.; Thomas, A. G.; Muryn, C. Photoelectron Spectroscopy Study of Stoichiometric and Reduced Anatase TiO 2 (101) Surfaces: The Effect of Subsurface Defects on Water Adsorption at Near-Ambient Pressures. J. Phys. Chem. C 2015, 119 (24), 13682− 13690. (36) Thind, S. S.; Wu, G.; Tian, M.; Chen, A. Significant Enhancement in the Photocatalytic Activity of N, W Co-Doped TiO 2 Nanomaterials for Promising Environmental Applications. Nanotechnology 2012, 23 (47), 475706. (37) Kitano, M.; Funatsu, K.; Matsuoka, M.; Ueshima, M.; Anpo, M. Preparation of Nitrogen-Substituted TiO2 Thin Film Photocatalysts by the Radio Frequency Magnetron Sputtering Deposition Method and Their Photocatalytic Reactivity under Visible Light Irradiation. J. Phys. Chem. B 2006, 110 (50), 25266−25272. (38) Wong, M.-S.; Pang Chou, H.; Yang, T.-S. Reactively Sputtered N-Doped Titanium Oxide Films as Visible-Light Photocatalyst. Thin Solid Films 2006, 494 (1−2), 244−249. (39) Rumaiz, A. K.; Woicik, J. C.; Cockayne, E.; Lin, H. Y.; Jaffari, G. H.; Shah, S. I. Oxygen Vacancies in N Doped Anatase TiO[sub 2]: Experiment and First-Principles Calculations. Appl. Phys. Lett. 2009, 95 (26), 262111. (40) Mishra, T.; Mahato, M.; Aman, N.; Patel, J. N.; Sahu, R. K. A Mesoporous WN Co-Doped Titania Nanomaterial with Enhanced Photocatalytic Aqueous Nitrate Removal Activity under Visible Light. Catal. Sci. Technol. 2011, 1 (4), 609. (41) Li, J.; Xu, J.; Dai, W. L.; Li, H.; Fan, K. One-Pot Synthesis of Twist-like Helix Tungsten-Nitrogen-Codoped Titania Photocatalysts with Highly Improved Visible Light Activity in the Abatement of Phenol. Appl. Catal., B 2008, 82 (3−4), 233−243. (42) Ng, K. T.; Hercules, D. M. Studies of Nickel-Tungsten-Alumina Catalysts by X-Ray Photoelectron Spectroscopy. J. Phys. Chem. 1976, 80 (19), 2094−2102. (43) Sun, M.; Xu, N.; Cao, Y. W.; Yao, J. N.; Wang, E. G. Nanocrystalline Tungsten Oxide Thin Film: Preparation, Microstructure, and Photochromic Behavior. J. Mater. Res. 2000, 15 (4), 927−933. (44) Ahrenkiel, R. K.; Johnston, S. W. Interaction of Microwaves with Photoelectrons in Semiconductors. J. Vac. Sci. Technol. B Microelectron. Nanom. Struct. 2008, 26 (4), 1508.

(45) Savenije, T. J.; Huijser, A.; Vermeulen, M. J. W.; Katoh, R. Charge Carrier Dynamics in TiO2 Nanoparticles at Various Temperatures. Chem. Phys. Lett. 2008, 461 (1−3), 93−96. (46) Martin, S. T.; Herrmann, H.; Choi, W.; Hoffmann, M. R. TimeResolved Microwave Conductivity. Part 1.TiO 2 Photoreactivity and Size Quantization. J. Chem. Soc., Faraday Trans. 1994, 90 (21), 3315−3322. (47) Martin, S. T.; Herrmann, H.; Hoffmann, M. R. Time-Resolved Microwave Conductivity. Part 2.Quantum-Sized TiO 2 and the Effect of Adsorbates and Light Intensity on Charge-Carrier Dynamics. J. Chem. Soc., Faraday Trans. 1994, 90 (21), 3323−3330. (48) Nakajima, S.; Katoh, R. Time-Resolved Microwave Conductivity Study of Charge Carrier Dynamics in Commercially Available TiO 2 Photocatalysts. J. Mater. Chem. A 2015, 3 (30), 15466−15472. (49) Daghrir, R.; Drogui, P.; Ka, I.; El Khakani, M. A. Photoelectrocatalytic Degradation of Chlortetracycline Using Ti/TiO2 Nanostructured Electrodes Deposited by Means of a Pulsed Laser Deposition Process. J. Hazard. Mater. 2012, 199−200, 15−24. (50) Giroux, I.; Roy, N.; Lamontagne, C. Présence de Pesticides Dans l’Eau Souterraine En Milieu Agricole: Étude Pilote Du Bassin Versant de La Rivière Châteauguay. Can. Water Resour. J. 2010, 35 (4), 527−542. (51) Trentacoste, S. V.; Friedmann, a S.; Youker, R. T.; Breckenridge, C. B.; Zirkin, B. R. Atrazine Effects on Testosterone Levels and Androgen-Dependent Reproductive Organs in Peripubertal Male Rats. J. Androl. 2001, 22 (1), 142−148. (52) Komtchou, S.; Dirany, A.; Drogui, P.; Delegan, N.; El Khakani, M. A.; Robert, D.; Lafrance, P. Degradation of Atrazine in Aqueous Solution with Electrophotocatalytic Process Using TiO2−x Photoanode. Chemosphere 2016, 157, 79−88.

I

DOI: 10.1021/acs.jpcc.7b11266 J. Phys. Chem. C XXXX, XXX, XXX−XXX